18650s -- 2200 MWh & 3200 MWh in same build pack?
- Thread starter 55rebel
- Start date
Hillhater
100 TW
2 separate 48v , 13S packs ?55rebel said:Newbie here...
I'm hoping to build a battery pack for an e-bike using 195 2200 MWh batteries (13S15P), and 39 @ 3200 MWh (13S3P).
Q: Would this kosher?
If you connect them in series, for a 96v pack, you will only get to use the capacity of the 13s, 3p pack before you distroy them.
If you connect them in parallel you risk killing the 13s, 3p section......unless the discharge current is VERY low.
Use them as separate packs.
Hillhater
100 TW
Yes, that would be a good solution for 48v use.
Its sometimes nice to have a small , lighweight, low capacity battery option for those short trips
2kg vs 10kg for the big pack is quite a difference in ride experience....and quick to recharge !
Its sometimes nice to have a small , lighweight, low capacity battery option for those short trips
2kg vs 10kg for the big pack is quite a difference in ride experience....and quick to recharge !
This thread has some confusions in it.
The first is that there aren't any 18650 cells that have 2200 ro 3200 Mega Watt hours of capacity.
Probably not milli Watt hours either, as that would be a very low capacity cell.
I suspect you mean mAh, which is milli Amp hours.
The second is that paralleling a lower-capacity pack with a higher-capacity pack is a lot better idea than running them separately, as long as both of them are the same chemistry and number of cells, and are at the same state of charge when you parallel them.
If you can, parallel them at the cell-level, rather than just at the ends.
Either way it will ease the load on both packs, compared to running them separately.
If you run a small-capacity pack separately, it's a lot higher load on it, and it's a lot harder on it than if it was paralleled with the other pack.
Seriesing them *would* be a bad idea, however.
The first is that there aren't any 18650 cells that have 2200 ro 3200 Mega Watt hours of capacity.
Probably not milli Watt hours either, as that would be a very low capacity cell.
I suspect you mean mAh, which is milli Amp hours.
The second is that paralleling a lower-capacity pack with a higher-capacity pack is a lot better idea than running them separately, as long as both of them are the same chemistry and number of cells, and are at the same state of charge when you parallel them.
If you can, parallel them at the cell-level, rather than just at the ends.
Either way it will ease the load on both packs, compared to running them separately.
If you run a small-capacity pack separately, it's a lot higher load on it, and it's a lot harder on it than if it was paralleled with the other pack.
Seriesing them *would* be a bad idea, however.
Thanks for the education/correction on those batteries amberwolf. Like I said... "newbie here" 
I was under the impression that Hillhater knew what he was talking about when he said that I would ruin the 39 3200 mAh batteries (13s-3p) if ran them in parallel with the lower rating batteries (2200 mAh -- 13s-15p).
Am I missing something here?
Thanks
I was under the impression that Hillhater knew what he was talking about when he said that I would ruin the 39 3200 mAh batteries (13s-3p) if ran them in parallel with the lower rating batteries (2200 mAh -- 13s-15p).
Am I missing something here?
Thanks
I don't see a reason that would happen if they are in parallel.
It could happen in series (and even then, only if used beyond either the smallest packs' capacity, or the C-rate limitations of the lowest-capability pack).
The lowest internal-resistance (Ri) cell set would provide most of the current during high-draw situations. If that is also the largest portion of the pack, it'd be fine as-is. If it's the smallest portion of the pack, then those cells would be recharged by current flow from the higher-resistance larger portion of the pack as soon as the load lessens enough for current flow to be equalized between them.
But a problem could happen:
If load is always higher than the higher-Ri portion of the pack can handle, then those cells won't contribute much to the pack until the lower-Ri portion has supplied all the capacity it can down to the voltage the pack sags to at that load, at which point the voltage will begin sagging even more as the higher-Ri portion is forced to supply the current.
If the load is high enough and is continuous, it'll be pretty hard on the pack.
As long as the load is intermittent then the lower-Ri portion will continue to be recharged by the higher-Ri portion.
But it'd still be easier on it than using either portion alone, as long as you know the packs' limits and don't abuse them.
BTW, lower capacity doesn't equate with lower capability (C-rate) or higher Ri; you'd ahve to look those up on the manufacturer spec sheets for each model of cell you're using.
Also, if you don't know who made the cells, and they're not marked with their genuine manufacturer info, then you don't actually have any benchmarks to go by unless you test them all yourself to see what they really are. (counterfeit and junk cells are common)
So you need to know the load you're going to be putting on them with the bike. And you need to know what they are capable of, to be sure you can use them for that bike at that load.
It could happen in series (and even then, only if used beyond either the smallest packs' capacity, or the C-rate limitations of the lowest-capability pack).
The lowest internal-resistance (Ri) cell set would provide most of the current during high-draw situations. If that is also the largest portion of the pack, it'd be fine as-is. If it's the smallest portion of the pack, then those cells would be recharged by current flow from the higher-resistance larger portion of the pack as soon as the load lessens enough for current flow to be equalized between them.
But a problem could happen:
If load is always higher than the higher-Ri portion of the pack can handle, then those cells won't contribute much to the pack until the lower-Ri portion has supplied all the capacity it can down to the voltage the pack sags to at that load, at which point the voltage will begin sagging even more as the higher-Ri portion is forced to supply the current.
If the load is high enough and is continuous, it'll be pretty hard on the pack.
As long as the load is intermittent then the lower-Ri portion will continue to be recharged by the higher-Ri portion.
But it'd still be easier on it than using either portion alone, as long as you know the packs' limits and don't abuse them.
BTW, lower capacity doesn't equate with lower capability (C-rate) or higher Ri; you'd ahve to look those up on the manufacturer spec sheets for each model of cell you're using.
Also, if you don't know who made the cells, and they're not marked with their genuine manufacturer info, then you don't actually have any benchmarks to go by unless you test them all yourself to see what they really are. (counterfeit and junk cells are common)
So you need to know the load you're going to be putting on them with the bike. And you need to know what they are capable of, to be sure you can use them for that bike at that load.
amberwolf said:
Thanks for the great explanation amberwolf. Sorry for the belated responce... been out of town.
The cells I'm using are LG cells. rated at 2150 mAh according to their spec sheet, but according to those that are buying and testing them, they are more like 2200+.
I plan on using 196 of these 200 cells ( http://www.ebay.com/itm/10-LG-36V-4-4AH-BATTERY-PACK-18650-EBIKE-VAPE-POWERWALL-BATTERIES-200-CELLS-BMS/201916138726?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649 ), plus 28 of the 40 Panasonic NCR18650BE / MH12210 cells I have. I've decided to go with 14s-16p for a 52V (58.8V) battery pack, to power a 1000W brushless hub motor. Hoping to get at least 50 mi. range out of that battery pack.
My only other question (NOT) would be... how to calculate the amp-draw on each individual cell at max load, so I can determine the fuse wire requirements for those cells, and how often I should ballance charge them....if ever, after the initial balance charge ( https://www.youtube.com/watch?v=tbaeMWdDlq ).
Thanks!
Don't count on more than spec. Nice if you get it but if you count on it and it's not there, you have disappointment at best and problems at worst.55rebel said:
I'd be wary of them actually being what they claim (regardless of anything on the ebay pages or comments, which can all be faked). I've seen what appear at a glance to be the same pics on various ads, each of which claims them to be different brands, etc. I've not checked up on any of them or looked at any of the pics in detail, so you'd have to do that research if you are interested.
At least one version of these ads has been discussed here on ES, though I don't recall what the conclusions were, if any.
You'll need to determine what your estimated wh/mile is (under your specific riding and traffic conditions and terrain) to know what range you might actually get out of the pack.
At 16p, if they are ~2Ah cells, even if they are only good for 1C, you can still pull about 32A safely from them. At 52v average that's 1600W+, so teh pack will probably easily handle the 20A controller that a 1000w "48v" or "52v" kit would have on it.
Max load is determined by the controller. If it's a 20A controller, it'll probably pull as much as 25A+ at peak, but likely sustain only 20-22A. You'll have to measure it with a wattmeter once built to find out for sure.
So if it's a 20A load, and you have 16 cells in parallel, then you divide 20 by 16. Substitute whatever A number your controller actually uses.
Depends entirely on how quickly they get out of balance in use. That's something you'll have to check, either with a voltmeter checking each cell group at full charge and at empty both before and during recharge, or with cell-monitors (like Celllogs or Battery Medics or any of the zillion other RC-hobby cell-checkers), or with a BMS that does all this for you.
If running without a BMS make sure your controller's LVC is well above the "empty" point of the pack, or you risk overdischarging low cell groups.
I can't watch the video; YT says it doesn't exist.
Sorry about that; I didn't capture the whole url. If your still interested... https://www.youtube.com/watch?v=tbaeMWdDlqYamberwolf said:
It's basically jehugarcia explaining why he doesn't balance the battery pack in his converted 1962 VW panel bus and hasn't needed to in over 2 yrs. Also, explains the different lipo chemistries and why the need to balance, or not, with demonstrations. Very edifying.
I have the imax B6, and will be testing ALL the cells. Haven't started the prosses yet, still disassembling those packs. Will update.amberwolf said:
Again, thanks for all the great info.
BTW, are you originally from Phoenix? I was born and pretty much raised here in Phoenix. So, I'm a bonafide "Desert Rat". 8)
FWIW, it isn't really the chemistry itself, it's the quality of the cells, and how hard they are pushed.55rebel said:
RC LiPo, for instance, is generally a low-quality manufacturing and probably little or no QC product. So many of them require cell-level balancing on every cycle, when deeply discharged or used at high rates.
The EIG NMC cells I use are high-quality with good QC, and so far I haven't had to do balancing on any of them, though I have had one traction-pack cell begin to have lowered capacity and higher Ri after several years, and relegated it to lower-usage lighting-pack duty.
Some A123 pouch cells I have seem to stay balanced, though I haven't checked it manually in a long time (it has a BMS, but when first acquired used I did a fair bit of such testing to be sure it was ok, and also just to contribute to the ES knowledgebase).
Some cheap 18650 LiFePO4 cells I had (Vpower/CammyCC) had to be balanced every time, regardless of discharge level.
Some better 18650 Lithium cells I have in a pack that's underspec'd for what it's rated to do stay balanced ok as long as it's not used up to it's ratings, or discharged too deeply, though because it can't handle much load I don't really use it too often, as my uses are hard on batteries.
You'll probably see similar findings in various battery/cell test/troubleshooting/repair threads around ES.
Been here more than 30 years, and lived here a few years as a kid, too, with a few years stay in Texas farm country between those, but born on the Texas coast and then a few years on the California coast before my first stint here.
But it's my home now, and not going anywhere if I have a choice about it.
BTW, if you put your location in your profile, then whenever you post questions about vendors, repairs, etc., people can better respond with relevant info.
Similar threads
